Green Bank, West Virginia

October 1, 1993

TO: 140-ft Observers

FROM: Ronald J. Maddalena

SUBJECT: Data Archive

Traditionally, the disk drives attached to the control computer at the 140-ft have been used to store autocorrelator and continuum data until they could be archived to tape. During the last year we have noticed that the disk and tape drives, because of their age and the lack of spare parts, are having more failures and are becoming unreliable. I have recently initiated a way of archiving data that bypasses the Modcomp disk and tape drives.

The new method of archiving involves transferring autocorrelator and continuum data, as they are taken, directly to the analysis Sun workstation where they are stored on disk. I store two identical copies of the data on different disks so that, in case a disk crashes, we will not have lost any data. About once a week, the data are copied from the Sun's disk to Exabyte tape in exactly the same tape format as we used to write with the control Modcomp. The tape is then stored as part of the permanent archive.

During three months of tests, the new archive scheme has proven to be at least as reliable as the traditional mechanism. Furthermore, the new scheme has many advantages over the old. For example:

If observers want, I can easily provide software that will search through the archive data base for specific observations.

As of September 15, we have turned off the old data archive mechanism and are relying completely on the new mechanism.